CCL:G: Problem with Frequency Calculation Gaussian 03

 Sent to CCL by: "Adrian  Villlegas" [adriano-$-eps.mcgill.ca]
 Hello everyone,
 I was wondering if anybody can provide some help on the way to perform
 succesfully a frequency calculation using Gaussian 03.
 I have tried several times to run a frequency calculation (on a Pentium IV
 machine) of a succesfully optimized ionrganic cluster - Hartree-Fock-6-31G(d,p)-
 involving 50 atoms. The geometry optimization required the use of the
 "modredundant" function in order to freeze some of the atoms in the
 cluster.
 I have run the frequency calculation in two modes: i) inlcuding the modredundant
 function as in I did in the geometry optimisatin and ii) by unlocking all atoms
 in the cluster (modredundant option not considered), in both cases I used the
 same basis set as used for the geometry optimization: 6-31G(d,p). Thus far, the
 calculation runs until a point where the program crashes and a: "severe
 error message  2070" is displayed
 Below I have copy-pasted the last section of the incompleted calculation as
 displayed in the output file:
 The electronic state of the initial guess is 1-A.
  Requested convergence on RMS density matrix=1.00D-08 within 128 cycles.
  Requested convergence on MAX density matrix=1.00D-06.
  Requested convergence on             energy=1.00D-06.
  No special actions if energy rises.
  Restarting incremental Fock formation.
  SCF Done:  E(RHF) =  -8605.99834357     A.U. after   21 cycles
              Convg  =    0.5842D-08             -V/T =  2.0012
              S**2   =   0.0000
  Range of M.O.s used for correlation:     1   797
  NBasis=   797 NAE=   235 NBE=   235 NFC=     0 NFV=     0
  NROrb=    797 NOA=   235 NOB=   235 NVA=   562 NVB=   562
           Differentiating once with respect to electric field.
                 with respect to dipole field.
           Electric field/nuclear overlap derivatives assumed to be zero.
  Integrals replicated using symmetry in FoFDir.
  MinBra= 0 MaxBra= 2 Meth= 1.
  IRaf=       0 NMat=   3 IRICut=       1 DoRegI=T DoRafI=F ISym2E= 2 JSym2E=2.
           There are   3 degrees of freedom in the 1st order CPHF.
      3 vectors were produced by pass  0.
  AX will form   3 AO Fock derivatives at one time.
      3 vectors were produced by pass  1.
      3 vectors were produced by pass  2.
      3 vectors were produced by pass  3.
      3 vectors were produced by pass  4.
      3 vectors were produced by pass  5.
      3 vectors were produced by pass  6.
      3 vectors were produced by pass  7.
      3 vectors were produced by pass  8.
      3 vectors were produced by pass  9.
      3 vectors were produced by pass 10.
      3 vectors were produced by pass 11.
      3 vectors were produced by pass 12.
      3 vectors were produced by pass 13.
      3 vectors were produced by pass 14.
      3 vectors were produced by pass 15.
      3 vectors were produced by pass 16.
      3 vectors were produced by pass 17.
  Inv2:  IOpt= 1 Iter= 1 AM= 3.15D-15 Conv= 1.00D-12.
  Inverted reduced A of dimension   54 with in-core refinement.
  End of Minotr Frequency-dependent properties file   721 does not exist.
  Symmetrizing basis deriv contribution to polar:
  IMax=3 JMax=2 DiffMx= 0.00D+00
  G2DrvN: will do   18 centers at a time, making    3 passes doing MaxLOS=2.
  FoFDir/FoFCou used for L=0 through L=2.
           Differentiating once with respect to electric field.
                 with respect to dipole field.
           Differentiating once with respect to nuclear coordinates.
  Integrals replicated using symmetry in FoFDir.
  MinBra= 0 MaxBra= 2 Meth= 1.
  IRaf=       0 NMat=  52 IRICut=      52 DoRegI=T DoRafI=T ISym2E= 2 JSym2E=2.
  Raff turned off since only 49.75% of shell-pairs survive.
           There are 156 degrees of freedom in the 1st order CPHF.
    153 vectors were produced by pass  0.
  AX will form  51 AO Fock derivatives at one time.
    153 vectors were produced by pass  1.
  writwa
 Thanks a lot for your feedback
 Adrin